Breaking out of the OLAP Lock Box
June 20, 2008
For many years OLAP cubes have been the mainstay of the most successful business intelligence offerings. OLAP offered almost instant access to the key business measures at the key ‘pulse points’ which determine the success or failure of your business, these can then be intuitively analysed according to the ‘business dimensions’ defined.
OLAP offered speed of response and business friendly information, when compared to ‘query by example’ offerings which were little more than simplified SQL the ability to analyse business performance and drill further into the detail to determine the root cause of good or bad performance at the macro level could add immediate value to any business.
The massive value add of OLAP technology encouraged a proprietary approach to the storage and access of data which inhibited the spread of OLAP based solutions, as a result these became the preserve of specialists rather than becoming fully embedded in the business.
The development and promotion of the OLE DB for OLAP standard represented the first step of many required to open up OLAP to allow it to become simply another source of data to be consumed by the myriad of available tools. However adoption was not widespread and many end user tools remain firmly based on the ODBC standards.
The convergence of Business Intelligence data and XML, which is expressed in the XML for Analysis standard (XML/A), makes the prospect of seamlessly integrating OLAP data into standard business reporting a viable prospect.
Conventionally integrating the information held in OLAP cubes, even those which conform to the existing standards, requires specialised tools or advanced skills (MDX, COM etc.). XML/A offers a way to quickly and easily integrate OLAP data from any of these providers into your existing reporting environment:
- Microsoft SQL Server Analysis Services (MSAS)
- SAP BW Infocubes
- Hyperion’s Essbase
The key benefit of OLAP analysis to a business is the ability to quickly analyse information over time and according to a set of pre-defined business dimensions. This delivers focused value added information directly to business users.
Evolving standards may soon make the original promise of OLAP a reality for all.
All Trademarks are the properties of their respective owners.
The Evolving BI Landscape
May 21, 2008
The BI Market is consolidating and continuing to evolve at considerable speed. The recent wave of acquisitions has resulted in 3 broad market categories:
1. Vertical Applications
2. Niche Value/Best of Breed Applications
3. Technology Platforms
The Vertical Applications market is evenly split between Oracle and SAP, both companies have acquired a multiplicity of BI tools and technologies and are focused on extending the value of the core applications offering.
Examples of niche/best of breed applications would be the newly emerging Qlikview or the Logi9 BI platform from LogiXML in the Reporting and Analytics space or Teradata/Netezza/Vertica in the high performance data warehouse space. Some of these products are new some have been around for a while and serve a particular niche well.
The third category Technology Platforms is dominated by IBM and Microsoft, both of these companies provide a complete ‘Information Server’ platform which can be used to source, package and deliver information to a multiplicity of different users.
Evolving in parallel to these 3 market categories is the open source BI movement, however despite offering a wide range of functionality at a compelling cost in terms of software acquisition. The costs which will be incurred over the full lifecycle of the solution are still far from understood.
However there appears to be a general reluctance among users of these platforms to deliver the end to end solution entirely from one vendor. The acquisition of the de-facto standard in enterprise reporting, Cognos 8, by IBM will definitely put this to the test.
Microsoft actively promotes both it’s own desktop/intranet focused tools and provides the platform for many niche players. An emerging alternative to the combination of IBM and Cognos in the coming year, will be the combination of the Microsoft ‘Information Server’ platform with the unified Business Intelligence offering from LogiXML, Logi9.
The successful entrance of Logi into this market was acknowledged by Gartner in the 2008 BI Quadrant.
Watch this space ………………….
The Data Vault
May 17, 2008
The Data Vault architecture offers a unique solution to business problems and technical problems alike. It is focused squarely at the data integration efforts across the enterprise and is built from solid foundational concepts. A key to understanding the Data Vault is understanding the business. Once the business is mapped out and the practitioner has a firm grasp on how the business operates, then the process of building the Data Vault can commence.
The Data Vault has many benefits which are produced as a by-product of the basic engineering. Sticking to the Data Vault foundational rules and standards will help get any integration project off the ground quickly and easily.
It is very easy to convert both 3rd normal form and Star Schema to Data Vault model architecture, here we show how to convert from 3rd normal form.
-
Data Vault Benefits
1. Manage and Enforce Compliance to Sarbanes-Oxley, HIPPA, and BASIL II in your Enterprise Data Warehouse
2. Spot business problems that were never visible previously
3. Rapidly Reduce business cycle time for implementing changes
4. Merge new business units into the organization rapidly
5. Rapid ROI and Delivery of information to new Star Schemas
6. Consolidate disparate data stores., ie: Master Data Management
7. Implement and Deploy SOA.
8. Scale to hundreds of Terabytes or Petabytes
9. SEI CMM Level 5 compliant (Repeatable, consistent, redundant architecture)
10.Trace all data back to the source systems
Definition: The Data Vault is a detail oriented, historical tracking and uniquely linked set of normalized tables that support one or more functional areas of business. It is a hybrid approach encompassing the best of breed between 3rd normal form (3NF) and star schema. The design is flexible, scalable, consistent and adaptable to the needs of the enterprise. It is a data model that is architected specifically to meet the needs of today’s enterprise data warehouses.
The Data Vault concept was pioneered by Dan Lindstedt
